Papers › Intriguing Findings of Frequency Selection for Image Deblurring

Intriguing Findings of Frequency Selection for Image Deblurring

23 Nov 2021arXiv:2111.11745archive 2025-07-28

Xintian Mao, Yiming Liu, Fengze Liu, Qingli Li, Wei Shen, Yan Wang

Blur was naturally analyzed in the frequency domain, by estimating the latent sharp image and the blur kernel given a blurry image. Recent progress on image deblurring always designs end-to-end architectures and aims at learning the difference between blurry and sharp image pairs from pixel-level, which inevitably overlooks the importance of blur kernels. This paper reveals an intriguing phenomenon that simply applying ReLU operation on the frequency domain of a blur image followed by inverse Fourier transform, i.e., frequency selection, provides faithful information about the blur pattern (e.g., the blur direction and blur level, implicitly shows the kernel pattern). Based on this observation, we attempt to leverage kernel-level information for image deblurring networks by inserting Fourier transform, ReLU operation, and inverse Fourier transform to the standard ResBlock. 1x1 convolution is further added to let the network modulate flexible thresholds for frequency selection. We term our newly built block as Res FFT-ReLU Block, which takes advantages of both kernel-level and pixel-level features via learning frequency-spatial dual-domain representations. Extensive experiments are conducted to acquire a thorough analysis on the insights of the method. Moreover, after plugging the proposed block into NAFNet, we can achieve 33.85 dB in PSNR on GoPro dataset. Our method noticeably improves backbone architectures without introducing many parameters, while maintaining low computational complexity. Code is available at https://github.com/DeepMed-Lab/DeepRFT-AAAI2023.

PaperPDFCode

In Syntology View this paper on Syntology, its page in Syntology's graph. That page lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

deepmed-lab/deeprft-aaai2023 officialmentioned in papermentioned on GitHubpytorchNOASSERTIONSyntology: not harvested report
invokerer/deeprft officialmentioned in papermentioned on GitHubpytorchNOASSERTIONSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to AdaRevD: Adaptive Patch Exiting Reversible Decoder Pushes the Limit of Image Deblurring (arXiv:2406.09135; that paper's own run record: 36 ran (of which 16 constructed an object rather than computing a result; 27 with no instrument failure: 5 honoured, 2 violated, 20 with no contract checked; 9 where Syntology's instrument failed) · 14 unverified); nothing checks that they implement this paper's method. report
INVOKERer/AdaRevD mentioned on GitHubpytorchNOASSERTIONSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to AdaRevD: Adaptive Patch Exiting Reversible Decoder Pushes the Limit of Image Deblurring (arXiv:2406.09135; that paper's own run record: 36 ran (of which 16 constructed an object rather than computing a result; 27 with no instrument failure: 5 honoured, 2 violated, 20 with no contract checked; 9 where Syntology's instrument failed) · 14 unverified); nothing checks that they implement this paper's method. report
INVOKERer/LoFormer mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to AdaRevD: Adaptive Patch Exiting Reversible Decoder Pushes the Limit of Image Deblurring (arXiv:2406.09135; that paper's own run record: 36 ran (of which 16 constructed an object rather than computing a result; 27 with no instrument failure: 5 honoured, 2 violated, 20 with no contract checked; 9 where Syntology's instrument failed) · 14 unverified); nothing checks that they implement this paper's method. report
deepmed-lab-ecnu/single-image-deblur mentioned on GitHubpytorchSyntology: not harvested report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

DeblurringImage DeblurringImage Defocus Deblurring

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Deblurring GoPro DeepRFT+ PSNR 33.52 #18 of 56 Archive leaderboard report
Deblurring GoPro DeepRFT+ SSIM 0.965 #18 of 56 Archive leaderboard report
Deblurring HIDE (trained on GOPRO) DeepRFT+ PSNR (sRGB) 31.66 #8 of 26 Archive leaderboard report
Deblurring HIDE (trained on GOPRO) DeepRFT+ SSIM (sRGB) 0.946 #8 of 26 Archive leaderboard report
Deblurring MSU BASED Deeprft (GoPro) ERQAv2.0 0.74323 #6 of 11 Archive leaderboard report
Deblurring MSU BASED Deeprft (GoPro) LPIPS 0.08326 #6 of 11 Archive leaderboard report
Deblurring MSU BASED Deeprft (GoPro) PSNR 31.57612 #6 of 11 Archive leaderboard report
Deblurring MSU BASED Deeprft (GoPro) SSIM 0.94484 #6 of 11 Archive leaderboard report
Deblurring MSU BASED Deeprft (GoPro) Subjective 0.5354 #6 of 11 Archive leaderboard report
Deblurring MSU BASED Deeprft (GoPro) VMAF 66.55057 #6 of 11 Archive leaderboard report
Deblurring MSU BASED Deeprft (REDS) ERQAv2.0 0.74339 #7 of 11 Archive leaderboard report
Deblurring MSU BASED Deeprft (REDS) LPIPS 0.08139 #7 of 11 Archive leaderboard report
Deblurring MSU BASED Deeprft (REDS) PSNR 31.32349 #7 of 11 Archive leaderboard report
Deblurring MSU BASED Deeprft (REDS) SSIM 0.94479 #7 of 11 Archive leaderboard report
Deblurring MSU BASED Deeprft (REDS) Subjective 0.4622 #7 of 11 Archive leaderboard report
Deblurring MSU BASED Deeprft (REDS) VMAF 66.46811 #7 of 11 Archive leaderboard report
Deblurring RealBlur-J DeepRFT+ PSNR (sRGB) 32.63 #9 of 17 Archive leaderboard report
Deblurring RealBlur-J DeepRFT+ SSIM (sRGB) 0.933 #9 of 17 Archive leaderboard report
Deblurring RealBlur-J (trained on GoPro) DeepRFT+ PSNR (sRGB) 28.88 #7 of 15 Archive leaderboard report
Deblurring RealBlur-J (trained on GoPro) DeepRFT+ SSIM (sRGB) 0.880 #7 of 15 Archive leaderboard report
Deblurring RealBlur-R DeepRFT+ PSNR (sRGB) 40.01 #9 of 17 Archive leaderboard report
Deblurring RealBlur-R DeepRFT+ SSIM (sRGB) 0.973 #9 of 17 Archive leaderboard report
Deblurring RealBlur-R (trained on GoPro) DeepRFT PSNR (sRGB) 36.11 #5 of 19 Archive leaderboard report
Deblurring RealBlur-R (trained on GoPro) DeepRFT SSIM (sRGB) 0.955 #5 of 19 Archive leaderboard report
Image Deblurring GoPro DeepRFT+ PSNR 33.52 #21 of 55 Archive leaderboard report
Image Deblurring GoPro DeepRFT+ SSIM 0.965 #21 of 55 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

Convolution

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections